2.0 Analysis 2.1 Introduction Once the vessel was salvaged, a thorough examination was conducted to determine her seaworthiness, watertight integrity and stability under various deck loading conditions. 2.2 Factors Contributing to the Ingress of Water The ineffectiveness of the under-deck scupper pipe valves and the lack of gaskets around the perimeter of the main deck hatch coamings, in conjunction with the low freeboard and after trim, are considered to have been the principal factors contributing to the ingress of water. Because the main deck plating in way of the lazaret was holed, any water shipped and/or accumulated on the deck would have found its way into the lazaret. Furthermore, the repeated immersion of the ship-side discharge by wave motions while the pump was inoperative would have permitted seawater to enter the lazaret by way of the two holes in the flexible discharge hose, and caused the vessel's after trim to increase. 2.3 Flooding/Sinking Sequence As the fish catch was brought on board from the moored trap nets, the vessel's freeboard was gradually reduced and her after trim increased. Because the trap nets were attached to the vessel's side, her normal buoyant response to the sea motions was restricted. Consequently, for about two and a half hours, while the vessel was tied up at one trap net and then the other, the prevailing waves of 300 to 450 mm regularly submerged the lazaret under-deck scupper pipes, and intermittently immersed the main deck and transom freeing ports, the lazaret bilge pump discharge and the small hole in the main deck plating. While the vessel remained stationary, the low static pressure head generated by such slight immersion was insufficient to firmly close the swing check valves in the lazaret hatch coaming under-deck scuppers. Consequently, the shipped water regularly bypassed the valves and flowed inboard, where the absence of a watertight gasket in the recessed hatch coaming allowed such water to enter and accumulate in the lazaret. The situation became steadily worse as the flooding continued. The marked after trim and the extent of main deck immersion observed shortly before the vessel sank are confirmed by post-casualty calculations which clearly show the vessel to have been in a highly vulnerable condition. When the lazaret was nearly half-flooded and some 3.5 m of the aftermost main deck immersed, about 30 per cent of the intact waterplane area was lost. Consequently, the centre of flotation of the remaining intact waterplane shifted forward of midships, causing a drastic reduction in longitudinal stability such that relatively small increases in trimming moment would have caused the marked after trims reported by the survivors. As speed was reduced and the vessel finally stopped, the venturi suction effect of the cowlings at the outboard ends of all the under-deck scuppers and the main deck freeing ports was lost. Consequently, as the flooding continued, the non-watertight boundaries of the lazaret hatch became more deeply immersed causing downflooding from the main deck. Subsequent immersion of the engine-room hatch accelerated the rate of downflooding, causing the after trim to suddenly increase and the vessel to lose reserve buoyancy and sink quickly by the stern. As the vessel was settling in the water, the totes retained some buoyancy and remained afloat, such that those nearest the navigation station became trapped under the overhanging steering canopy top as the vessel sank stern first. 2.4 Fish Totes - Hazard Associated with Stern-first Sinking Although the totes weighed approximately 550 kg when loaded with fish/ice/water, they still retained some buoyancy. All the persons on board the vessel were positioned forward of the totes and became trapped/pinned by the totes; only some managed to free themselves. This would indicate that the hazard posed by the fish totes when the vessel is involved in a stern-fist sinking was not fully appreciated. 2.5 Stability, Freeboard and Trim Analysis Post-casualty analysis of the vessel's stability before and at the time of the sinking is based on hydrostatic data generated from hull measurements taken from the salvaged vessel at the owner's premises, and lightship characteristics derived from an inclining experiment subsequently carried out at a nearby marina. At the time of the occurrence, the MISS STEPHANIE II was outfitted for trap net fishing operations and was fully fuelled with approximately 400 gallons of diesel oil on departure. Reportedly, 10 insulated fish tote boxes, 4 of which were filled with flaked ice for catch preservation, together with 20 fish sorting trays, were loaded on the main deck. The totes were unsecured and symmetrically located about the vessel's centre line, with the aftermost arranged approximately 800 mm forward of the transom. The trim and stability analysis focuses on the progressive increase in after trim and the related freeboard reduction during the loading of the fish catch, and addresses the most likely flooding sequence consistent with the findings of the post-casualty hull inspections and survivors' reports of the sinking. Because the vessel had no draught marks, departure draughts were not recorded; however, post-casualty calculations based on the reported loading details show that, after due allowance for the designed rake of keel, she was trimmed 225 mm by the stern. Furthermore, the calculated freeboard of 200 mm at the transom is consistent with that observed by the crew shortly before sailing from Grand Bend. Calculations also show that, when the initial 1,363.6 kg of catch was loaded and stowed with some flaked ice and water in tote boxes, the after trim increased to 320 mm and the main deck freeboard at the transom was reduced to 95 mm. Concurrently, the aftermost hatch coaming under-deck scupper pipe discharge and the lazaret bilge pump discharge were respectively 40 mm and 115 mm above water which, in the prevailing sea conditions, made both vulnerable to shipping water and initiating the flooding. Calculations show that, when an additional 909 kg of fish was loaded at the second trap net, the freeboard at the transom was reduced to about 50 mm, the aftermost under-deck scupper discharges were at the trimmed waterline, and the bilge pump discharge only 75 mm above water, conditions in which the shipping and retention of water on deck and the accumulation of flood water in the lazaret would undoubtedly occur. A further series of loading conditions with 25 per cent, 50 per cent, 75 per cent and 100 per cent flooding in the lazaret show that, while the after trim increased and the after end of the main deck was progressively submerged, the vessel still retained initial transverse and longitudinal stability, and would have remained afloat, but only in ideal static sea conditions. When the lazaret was approximately half-filled, its non-watertight hatchway became submerged and the resultant sudden onset of downflooding accelerated the rate at which the vessel settled and trimmed by the stern. As the flooding continued, the under-deck scupper discharges in the engine-room recessed hatch coamings also became immersed which, due to the vessel's motion and the prevailing seas, resulted in the progressive flooding and sudden downflooding through the non-watertight hatch, culminating in the loss of reserve buoyancy and stern-first sinking reported by the survivors. 3.0 Conclusions 3.1 Findings Pre-sailing testing of the after lazaret bilge pump would have shown that it was inoperative due to a broken fuse. The non-return valves in the lazaret hatch coaming under-deck scuppers did not prevent the entry of water when only slightly submerged. None of the main deck hatch coamings were fitted with watertight gaskets. The watertight integrity of the shell plating, rudder trunk, propeller shaft stern-tube gland and other through-hull fittings was intact. The loaded vessel's after trim, freeboard and sea motions were such that water initially entered and accumulated in the lazaret by way of intermittently immersed apertures in the afterdeck plating, bilge pump discharge pipe and the hatch coaming which was not fitted with watertight gaskets. The principal lazaret flooding was due to the repeated and prolonged immersion of the under-deck scuppers and hatch coaming which was not fitted with watertight gaskets. The stern-first sinking was due to the loss of reserve buoyancy and longitudinal stability caused by the sudden progressive flooding of the engine-room, by way of the under- deck scuppers and hatch coaming not fitted with watertight gaskets, when the vessel stopped. The master was not aware of the extent of the stern trim and water shipped on the afterdeck until informed by the Fisheries Officer. The rate of flooding in the engine-room and on to the main deck suddenly increased when the vessel's speed was reduced and the venturi suction effect at the outboard ends of the under-deck scuppers and main deck freeing ports was lost. The danger posed by the fish totes as the vessel sank stern first was not fully appreciated by the vessel's personnel. The child was fitted with an adult lifejacket because there was no child's lifejacket on board. The two guests were trapped because their escape was obstructed by buoyant fish totes which became caught under the steering position canopy as the vessel sank. The two guests drowned before rescue could be effected. 3.2 Causes The MISS STEPHANIE II was lost following an accumulation of water in the lazaret, which could not be discharged due to an inoperative bilge pump. Subsequent progressive flooding of the engine-room led to the loss of reserve buoyancy and longitudinal stability which resulted in the sudden stern-first sinking. 4.0 Safety Action 4.1 Action Taken 4.1.1 Watertight Integrity In November 1994, a Marine Safety Advisory (MSA) was forwarded to the Canadian Coast Guard (CCG). It apprised the CCG of the shortcomings in the seaworthiness and serviceability of the MISS STEPHANIE II. Since there are five other vessels of similar design and configuration operated in the area by the same owner, the MSA emphasized the need for evaluating the adequacy of the fittings and configuration of these vessels with a view to improving their watertight integrity. In response, the CCG removed the MISS STEPHANIE II and three other vessels from service for necessary modification. The fifth and sixth vessels have been fitted with sealed and secure hatch covers of improved design. All vessels of similar design will be modified to the satisfaction of the CCG before they can be returned to service.